{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/enhancing-compositional-generalization-via","title":"Enhancing Compositional Generalization via Compositional Feature Alignment","arxiv_id":"2402.02851","date":"2024-02-05","proceeding":null,"authors":["Haoxiang Wang","Haozhe Si","Huajie Shao","Han Zhao"],"abstract":"Real-world applications of machine learning models often confront data distribution shifts, wherein discrepancies exist between the training and test data distributions. In the common multi-domain multi-class setup, as the number of classes and domains scales up, it becomes infeasible to gather training data for every domain-class combination. This challenge naturally leads the quest for models with Compositional Generalization (CG) ability, where models can generalize to unseen domain-class combinations. To delve into the CG challenge, we develop CG-Bench, a suite of CG benchmarks derived from existing real-world image datasets, and observe that the prevalent pretraining-finetuning paradigm on foundational models, such as CLIP and DINOv2, struggles with the challenge. To address this challenge, we propose Compositional Feature Alignment (CFA), a simple two-stage finetuning technique that i) learns two orthogonal linear heads on a pretrained encoder with respect to class and domain labels, and ii) fine-tunes the encoder with the newly learned head frozen. We theoretically and empirically justify that CFA encourages compositional feature learning of pretrained models. We further conduct extensive experiments on CG-Bench for CLIP and DINOv2, two powerful pretrained vision foundation models. Experiment results show that CFA outperforms common finetuning techniques in compositional generalization, corroborating CFA's efficacy in compositional feature learning.","url_abs":"https://arxiv.org/abs/2402.02851v2","url_pdf":"https://arxiv.org/pdf/2402.02851v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"enhancing-compositional-generalization-via","repo_url":"https://github.com/haoxiang-wang/compositional-feature-alignment","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"clip","method_name":"CLIP"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2402.02851","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.02851"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/haoxiang-wang/compositional-feature-alignment","reach":null}],"summary":{"ran_fixture":2,"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":4,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"a76d114d2ce49ea7","entry":"cal_class_weight","repo":"haoxiang-wang/compositional-feature-alignment","repo_kind":"official","path":"linear_probe_exps.py","file_url":"https://github.com/haoxiang-wang/compositional-feature-alignment/blob/HEAD/linear_probe_exps.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a76d114d2ce49ea7"}},{"code_sha256_prefix":"f1b17b6c46d50c34","entry":"process_model","repo":"haoxiang-wang/compositional-feature-alignment","repo_kind":"official","path":"clip_finetune/finetune.py","file_url":"https://github.com/haoxiang-wang/compositional-feature-alignment/blob/HEAD/clip_finetune/finetune.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f1b17b6c46d50c34"}},{"code_sha256_prefix":"8e1e2c3f2604331a","entry":"remap_array","repo":"haoxiang-wang/compositional-feature-alignment","repo_kind":"official","path":"linear_probe_exps.py","file_url":"https://github.com/haoxiang-wang/compositional-feature-alignment/blob/HEAD/linear_probe_exps.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8e1e2c3f2604331a"}},{"code_sha256_prefix":"72e0fc5279b66ae3","entry":"parallelize_model","repo":"haoxiang-wang/compositional-feature-alignment","repo_kind":"official","path":"clip_finetune/finetune.py","file_url":"https://github.com/haoxiang-wang/compositional-feature-alignment/blob/HEAD/clip_finetune/finetune.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"72e0fc5279b66ae3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}